Small Cell Lung Cancer (SCLC), an aggressive form of lung cancer, represents about 20% of primary lung tumors and exhibits the most malignant phenotype of lung cancer (reviewed in Livingston, R. B., 1997, “Combined modality therapy of lung cancer”, Clin Cancer Res., 3:2638-47 and Gazdar, A. F., 1994, “The molecular and cellular basis of human lung cancer”, Anticancer Res., 13:261-68). Two thirds of patients diagnosed with SCLC are between the ages of 50 and 70, with a huge, although declining, male preponderance. There is no evidence for genetic predisposition linked with SCLC.
The predominant risk factor for SCLC is cigarette smoking. More than 95% of patients with SCLC are current or past smokers with direct correlation to both the number of cigarettes smoked per day and the duration of smoking. Ionizing radiation and occupational carcinogens are additional known risk factors for SCLC.
The natural history of SCLC differs from other types of lung cancer in the early and extensive spread of the disease. Seventy to eighty percent of the patients with SCLC have dissemination, occult or otherwise, at the time of presentation. Furthermore, SCLC has a rapid growth rate and the fastest doubling time of all of the types of lung cancer (25-160 days). The median survival time from the time of diagnosis is about 1.5 to 3 months for patients with extensive or limited disease, respectively.
SCLC usually presents as large, rapidly developing lesions arising from the centrally located tracheobronchial airways and invading the mediastinum (Stahel, R. A. et al., 1989, “Staging and prognostic factors in small cell lung cancer: A consensus report”, Lung Cancer, 5:119-26). Typically, patients present with a cough or dyspnea, wheezing, and/or chest pain. Weight loss, fatigue and anorexia occur in up to one third of the patients. At the time of diagnosis, two thirds of the patients with SCLC have one or more clinically detectable distant metastases, including bone (30%), liver (25%), bone marrow (20%) and the central nervous system (10%). In a screening setting, sensitivity of X-ray ranges from 45-50%, sputum cytology from 25-30% and their combination amounts to about 60-70% specificity of positive diagnosis.
SCLC is a form of lung cancer characterized by a neuroendocrine phenotype. This is evidenced by the presence of neurosecretory granules visible ultrastructurally in the cytoplasm. These granules contain peptide hormones such as ADH, gastrin releasing peptide and neuromedin. The neural cell adhesion molecule is found on the cell surface. Other cell surface antigens are also linked with other forms of lung cancer. Due to their low specificity and sensitivity, tumor markers, like neuron-specific enolase (NSE), creatinin kinase BB, or neuro-endocrinological markers, are not useful in the diagnostic phase of SCLC disease. The above tumor markers are elevated in about 60-65% of cases at the time of diagnosis of SCLC and correlate with tumor bulk. The usefulness of “relapse diagnosis anticipation”, i.e., the attempt to correlate the level of tumor markers with tumor progression and disease, is marginal since salvage treatments are virtually non-existent.
Since most SCLC patients are not candidates for surgery, the standard treatment for SCLC includes chemotherapy and radiotherapy in stages I-IV. SCLC is highly sensitive to initial chemotherapy. In spite of this early responsiveness, residual cells inflict a fatal relapse in most patients due to a re-emergence of chemoresistant variants. Consequently, since the cure rate is extremely low for patients with extensive SCLC disease, treatment must be considered palliative. For patients with relapsed, progressive disease, chemotherapy seldom shows clinical effectiveness or provides a lasting response.
More than 100 years ago, Paul Ehrlich proposed the use of antibodies as “magic bullets” to deliver toxins to cancer cells. The potential of targeted immunotherapy has since attracted the attention of generations of investigators. In 1975, with the development of the technology for producing monoclonal antibodies (MoAbs), (G. Kohler and C. Milstein, 1975, Nature, 256:495-497), it seemed that successful antibody therapy was imminent. However, early trials with monoclonal antibodies revealed significant obstacles to their use in cancer therapy. For example, immune rejection of murine monoclonal antibodies constituted the primary hurdle for making antibody therapy an effective and successful therapeutic. In addition, disappointingly low levels of cytotoxicity were reported during initial clinical experience (L. W. Kwak et al., 1995, Clinical applications of monoclonal antibodies, In: Biologic Therapy of Cancer, Eds. V. T. DeVita, Jr., S. Hellman and S. A. Rosenberg, 2nd Ed., JB Lippincott Co., Philadelphia, Pa., pp. 553-565).
Experience to date suggests that only a small fraction of injected antibody actually reaches a tumor (R. A. Miller et al., 1981, Lancet, ii:226-230). To maximize antibody binding to target molecules, an ideal antibody for cancer therapy should have a high affinity for its antigen (A. Hekman et al., 1991, Cancer Immunol. Immunother., 32:364-372). In addition, an effective unconjugated antibody should work synergistically with the host's immune system effector mechanisms. Therapeutic antibodies that induce effector mechanisms such as antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytolysis (T. A. Waldman et al., 1994, Ann. Oncol., 5 Suppl. 1:13-17) have the potential to provide targeted cancer therapy that is safe and effective without the use of potentially harmful conjugates such as toxins or radionuclides.
Nearly all monoclonal antibodies recognizing antigens on human cancer cells also bind to normal human cells expressing the same antigen (J. G. Jurcic et al., 1996, Cancer Chemotherapy and Biological Response Modifiers Annual, Eds. H. M. Pinedo, D. L. Longo and B. A. Chabner, pp. 168-188). This cross-reactivity potentially compromises therapeutic effectiveness and raises issues of toxicity, leading to the continued interest in defining antigenic targets that are unique to tumor cells.
Human SCLC is considered to be a feasible target for immunotherapy using radiolabeled monoclonal antibodies (J. Zeuthen and A. J. Vangsted, 1993, Acta Oncologica, 32:845-51; Y. Olabiran et al., 1994, Br. J. Cancer, 69:247-52; A. Smith et al., 1991, Oncology, 64:263-6; P. L. Beaumier et al., 1991, Cancer Res., 51:676-81; R. A. Stahel, 1989, Chest, 96:27S-29S; and M. Hosono et al., 1994, J. Nuclear Med., 35:296-300). However, SCLC-specific MoAbs isolated to date also react against one or more of neuroendocrine tissues, unrelated cells, or normal immune cells. Examples include MoAbs binding to neuronal cell adhesion molecule (NCAM), polysialic acid, cluster w4 antigen (CD24) (D. Jackson et al., 1992, Cancer Res., 52:5264-70), sialoglycoprotein antigen sGP 90-135, ganglioside GD2, and ganglioside fucosyl-GM1 (FucGM1) (J. Zeuthen and A. J. Vangsted, 1993, Acta Oncologica, 32:845-51). A MoAb (N901), which is specific to NCAM (CD56), was reported to bind to SCLC tumors and cell lines, as well as to cardiac muscles, natural killer (NK) cells, and peripheral nerves.
Understanding the tissue distribution of tumor-associated antigens on cancers and normal tissues is essential for the selection of targets for cancer immunotherapy. The majority of cancer antigens are self-antigens that are derived from and expressed by normal cells. Frequently, the cancer antigen is identical to the normal antigen although it is expressed at higher levels or endowed with a negligible mutation insufficient for its distinction from the self-antigen. One of the escape mechanisms of malignant cells from the immune system is their resemblance to their normal cell counterparts, thereby resulting in low visibility for the malignant cells by an individual's immune surveillance system.
The process that leads to the discovery of unique cancer antigens is long, tedious and elaborate. The screening process entails an exhaustive weeding out of antigens expressed on both cancer or tumor cells and normal tissues. The probes used for the discovery of such antigens are limited in their efficacy due to the fundamentally low immunogenicity of tumor antigens. In addition, serum samples with high titers from cancer patients are generally scarce. Utilization of such probes for screening is frequently thwarted due to the “identification” of multiple artifacts, or to false-positive hits.
Antigens are diverse in their immunogenicity, i.e. their ability to stimulate the immune response. When several antigens possessing distinct immunogenic properties co-exist in an antigenic preparation, their antigenic dominance regulates the intensity of the immune response to antigens. Therefore, the most robust immune response is developed against the strongest epitopes found in the antigenic preparation. Antigens with weaker immunogenicity will be disregarded, or the level of the immune response elicited against them will be negligible, due to the focus of the immune response on the stronger and more dominant epitopes. Hence, minor epitopes contained in an antigenic preparation (e.g., a vaccine) will be masked by the more immunogenic epitopes. Generating discerning MoAbs against a repertoire of minor epitopes present in an antigenic preparation (such as subcellular fractions of cancer cells) containing dominant epitopes has remained a challenge for many years.
However, in spite of the above-mentioned obstacles, the present invention provides new and specific monoclonal antibodies which are immunoreactive with SCLC cell surface antigens and which are useful in immunotherapy, diagnostic, imaging, monitoring and screening methodologies, to name a few. The present invention has solved the problem of generating myriads of non-specific antibodies by employing a technique of differential immunization, which involves, in part, tolerization with closely related antigens, e.g., on normal cells or on tumor cells exhibiting a similar phenotype, followed by immunization with the neoplastic cells of interest having unique cell surface antigen molecules. This differential/tolerization process allows the weeding out of B cells possessing undesired specificities from the entire pool of B cells, prior to the fusion of the B cells with immortalized cells to create hybridomas. Consequently, the frequency of hybridomas with the sought-after antigenic specificities is amplified in accordance with the present invention, and the entire screening process is greatly simplified.